J Korean Neurosurg Soc.  2025 Jan;68(1):7-18. 10.3340/jkns.2024.0100.

Optimizing Glioblastoma, IDH-wildtype Treatment Outcomes : A Radiomics and Support Vector Machine-Based Approach to Overall Survival Estimation

• Chong JK
• Jain P
• Prasad S
• Dubey NK
• Saxena S
• Lo WC

• KMID: 2562978
• DOI: http://doi.org/10.3340/jkns.2024.0100

Abstract

Objective
: Glioblastoma multiforme (GBM), particularly the isocitrate dehydrogenase (IDH)-wildtype type, represents a significant clinical challenge due to its aggressive nature and poor prognosis. Despite advancements in medical imaging and its modalities, survival rates have not improved significantly, demanding innovative treatment planning and outcome prediction approaches.
Methods
: This study utilizes a support vector machine (SVM) classifier using radiomics features to predict the overall survival (OS) of GBM, IDH-wildtype patients to short (<12 months) and long (≥12 months) survivors. A dataset comprising multi-parametric magnetic resonance imaging scans from 574 patients was analyzed. Radiomic features were extracted from T1, T2, fluid-attenuated inversion recovery, and T1 with gadolinium (T1GD) sequences. Low variance features were removed, and recursive feature elimination was used to select the most informative features. The SVM model was trained using a k-fold cross-validation approach. Furthermore, clinical parameters such as age, gender, and MGMT (O6-methylguanine-DNA methyltransferase) promoter methylation status were integrated to enhance prediction accuracy.
Results
: The model showed reasonable results in terms of cross-validated area under the curve of 0.84 (95% confidence interval, 0.80–0.90) with (p<0.001) effectively categorizing patients into short and long survivors. Log-rank test (chi-square statistics) analysis for the developed model was 0.00029 along with the 1.20 Cohen’s d effect size. Most importantly, clinical data integration further refined the survival estimates, providing a more fitted prediction that considers individual patient characteristics by Kaplan-Meier curve with p-value <0.0001.
Conclusion
: The proposed method significantly enhances the predictive accuracy of OS outcomes in GBM, IDH-wildtype patients. By integrating detailed imaging features with key clinical indicators, this model offers a robust tool for personalized treatment planning, potentially improving OS.

Keyword

Glioblastoma multiforme; Isocitrate dehydrogenase-wildtype; Overall survival; Radiomics & radiogenomics

Figure

• 

  Fig. 1. Different structural modalities of multi-parametric magnetic resonance imaging. T1 : T1-weighted, T2 : T2-weighted, FLAIR : fluid-attenuated inversion recovery, T1GD : T1 with gadolinium, Segmask : segmentation mask.

• 

  Fig. 2. A whole tumor with its subregions such as tumor core (TC) marked as 1, edema (ED) marked as 2, and enhancing tumor (ET) marked as 4 [29].

• 

  Fig. 3. Survival days on replication cohort.

• 

  Fig. 4. Overview of the support vector machine (SVM)-based radiomics workflow for predicting survival outcomes in glioblastoma, IDH-wildtype patients. T1 : T1-weighted, T2 : T2-weighted, FLAIR : fluid-attenuated inversion recovery, T1-CE : T1 contrast enhanced, MRI : magnetic resonance imaging, LVR : low variance removal, RFE : recursive feature elimination, MGMT : O6-methylguanine-DNA methyltransferase, KM : Kaplan-Meier, IDH : isocitrate dehydrogenase.

• 

  Fig. 5. Receiver operating characteristic (ROC) curve across 10-folds. OS : overall survival, AUC : area under the curve.

• 

  Fig. 6. Survival index on replication cohort.

• 

  Fig. 7. Survival analysis using Kaplan-Meier curve.

Reference

References

1. Afshar P, Mohammadi A, Plataniotis KN, Oikonomou A, Benali H. From handcrafted to deep-learning-based cancer radiomics: challenges and opportunities. IEEE Signal Process Mag. 36:132–160. 2019.

2. Agravat RR, Raval MS. Prediction of overall survival of brain tumor patients. TENCON 2019-2019 IEEE Region 10 Conference (TENCON); Kochi: India;2019. p. p31.

3. Bakas S, Sako C, Akbari H, Bilello M, Sotiras A, Shukla G, et al. Multi-parametric magnetic resonance imaging (mpMRI) scans for de novo Glioblastoma (GBM) patients from the University of Pennsylvania Health System (UPENN-GBM) (version 2) [Data set]. Available at : https://doi.org/10.7937/TCIA.709X-DN49.

4. Bakas S, Sako C, Akbari H, Bilello M, Sotiras A, Shukla G, et al. The University of Pennsylvania glioblastoma (UPenn-GBM) cohort: advanced MRI, clinical, genomics, & radiomics. Sci Data. 9:453. 2022.

5. Bezerra FE, Oliveira Neto GCD, Cervi GM, Francesconi Mazetto R, Faria AMD, Vido M, et al. Impacts of feature selection on predicting machine failures by machine learning algorithms. Appl Sci. 14:3337. 2024.

6. Cerono G, Melaiu O, Chicco D. Clinical feature ranking based on ensemble machine learning reveals top survival factors for glioblastoma multiforme. J Healthc Inform Res. 8:1–18. 2023.

7. Chato L, Chow E, Latifi S : Wavelet transform to improve accuracy of a prediction model for overall survival time of brain tumor patients based on MRI images. 2018 IEEE International Conference on Healthcare Informatics (ICHI); 2018, Jun 4-7; New York : USA, 2018. pp441-442.

8. Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, et al. The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository. J Digit Imaging. 26:1045–1057. 2013.

9. Das S, Bose S, Nayak GK, Saxena S. Deep learning-based ensemble model for brain tumor segmentation using multi-parametric MR scans. Open Comput Sci. 12:211–226. 2022.

10. Fathi Kazerooni A, Saxena S, Toorens E, Tu D, Bashyam V, Akbari H, et al. Clinical measures, radiomics, and genomics offer synergistic value in AI-based prediction of overall survival in patients with glioblastoma. Sci Rep. 12:8784. 2022.

11. Forghani R, Savadjiev P, Chatterjee A, Muthukrishnan N, Reinhold C, Forghani B. Radiomics and artificial intelligence for biomarker and prediction model development in oncology. Comput Struct Biotechnol J. 17:995–1008. 2019.

12. Galbraith K, Kumar A, Abdullah KG, Walker JM, Adams SH, Prior T, et al. Molecular correlates of long survival in IDH-wildtype glioblastoma cohorts. J Neuropathol Exp Neurol. 79:843–854. 2020.

13. Guyon I, Elisseeff A. An introduction to variable and feature selection. J Mach Learn Res. 3:1157–1182. 2003.

14. Hajal C, Le Roi B, Kamm RD, Maoz BM. Biology and models of the blood-brain barrier. Annu Rev Biomed Eng. 23:359–384. 2021.

15. Hajianfar G, Shiri I, Maleki H, Oveisi N, Haghparast A, Abdollahi H, et al. Noninvasive O6 methylguanine-DNA methyltransferase status prediction in glioblastoma multiforme cancer using magnetic resonance imaging radiomics features: univariate and multivariate radiogenomics analysis. World Neurosurg. 132:e140–e161. 2019.

16. Hastie T, Tibshirani R, Friedman JH : The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Standford : Springer, 2009.

17. Hermoza R, Maicas G, Nascimento JC, Carneiro G : Post-hoc overall survival time prediction from brain MRI. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI); Nice, France, 2021, pp1476-1480.

18. Kanderi T, Gupta V : Glioblastoma Multiforme in Ackley WB, Adolphe TS, Aeby TC, Aeddula NR, Agadi S, Agasthi P, et al. (eds) : StatPearls [Internet]. Treasure Island : StatPearls Publishing, 2022.

19. Kotecha R, Odia Y, Khosla AA, Ahluwalia MS. Key clinical principles in the management of glioblastoma. JCO Oncol Pract. 19:180–189. 2023.

20. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer. 48:441–446. 2012.

21. Li G, Li L, Li Y, Qian Z, Wu F, He Y, et al. An MRI radiomics approach to predict survival and tumour-infiltrating macrophages in gliomas. Brain. 145:1151–1161. 2022.

22. Liu W, Liang S, Qin X. Weighted p-norm distance t kernel SVM classification algorithm based on improved polarization. Sci Rep. 12:6197. 2022.

23. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK. WHO classification of tumors of the central nervous system (revised 4th edition). Lyon: IARC;2016. p. p8.

24. Mali SA, Ibrahim A, Woodruff HC, Andrearczyk V, Müller H, Primakov S, et al. Making radiomics more reproducible across scanner and imaging protocol variations: a review of harmonization methods. J Pers Med. 11:842. 2021.

25. Mesti T, Ocvirk J. Malignant gliomas: old and new systemic treatment approaches. Radiol Oncol. 50:129–138. 2016.

26. Mogensen UB, Ishwaran H, Gerds TA. Evaluating random forests for survival analysis using prediction error curves. J Stat Softw. 50:1–23. 2012.

27. Nti IK, Nyarko-Boateng O, Aning J. Performance of machine learning algorithms with different K values in K-fold crossvalidation. Int J Inf Technol Comput Sci. 6:61–71. 2021.

28. Saxena S, Agrawal A, Dash P, Jena B, Khanna NN, Paul S, et al. Prediction of O-6-methylguanine-DNA methyltransferase and overall survival of the patients suffering from glioblastoma using MRI-based hybrid radiomics signatures in machine and deep learning framework. Neural Comput Appl. 35:13647–13663. 2023.

29. Saxena S, Jena B, Mohapatra B, Gupta N, Kalra M, Scartozzi M, et al. Fused deep learning paradigm for the prediction of o6-methylguanine-DNA methyltransferase genotype in glioblastoma patients: a neuro-oncological investigation. Comput Biol Med. 153:106492. 2023.

30. Singh G, Manjila S, Sakla N, True A, Wardeh AH, Beig N, et al. Radiomics and radiogenomics in gliomas: a contemporary update. Br J Cancer. 125:641–657. 2021.

31. Siontis KC, Noseworthy PA, Attia ZI, Friedman PA. Artificial intelligence-enhanced electrocardiography in cardiovascular disease management. Nat Rev Cardiol. 18:465–478. 2021.

32. Song L, Liang Q, Chen H, Hu H, Luo Y, Luo Y. A new approach to optimize SVM for insulator state identification based on improved PSO Algorithm. Sensors (Basel). 23:272. 2023.

33. Tan Y, Mu W, Wang XC, Yang GQ, Gillies RJ, Zhang H. Improving survival prediction of high-grade glioma via machine learning techniques based on MRI radiomic, genetic and clinical risk factors. Eur J Radiol. 120:108609. 2019.

34. Tang W, Zhang H, Yu P, Kang H, Zhang R. MMMNA-net for overall survival time prediction of brain tumor patients. Annu Int Conf IEEE Eng Med Biol Soc. 2022:3805–3808. 2022.

35. Thakkar JP, Dolecek TA, Horbinski C, Ostrom QT, Lightner DD, BarnholtzSloan JS, et al. Epidemiologic and molecular prognostic review of glioblastoma. Cancer Epidemiol Biomarkers Prev. 23:1985–1996. 2014.

36. van Timmeren JE, Leijenaar RTH, van Elmpt W, Reymen B, Oberije C, Monshouwer R, et al. Survival prediction of non-small cell lung cancer patients using radiomics analyses of cone-beam CT images. Radiother Oncol. 123:363–369. 2017.

37. Wilson TA, Karajannis MA, Harter DH. Glioblastoma multiforme: state of the art and future therapeutics. Surg Neurol Int. 5:64. 2014.

38. Yogananda CGB, Shah BR, Nalawade SS, Murugesan GK, Yu FF, Pinho MC, et al. MRI-based deep-learning method for determining glioma MGMT promoter methylation status. AJNR Am J Neuroradiol. 42:845–852. 2021.